A systematic method for prediction uncertainty quantification and decomposition in foliar uptake models

This is the authors' abstract. We don't add key points for this paper.

Models that characterize the uptake of pesticides through the leaves can find applications to tackle many challenges of the present and of the future related to food security, such as in supporting the research of novel biocides with in silico studies that can reduce experimentation cost and time. However, to make sensible use of these models, their predictions must be carefully evaluated especially due to the intrinsic uncertainty arising from the modelling assumptions and to the biological system characteristics. In this paper different sources of uncertainty and their effect are studied, including uncertainty in the inputs, in model parameters and in the variance model used to replicate real experiments in silico, paving the way for reliable applications of computational tools in agrochemical research.